Social risk management through transitional labour markets - CiteSeerX

Socio-Economic Review (2006) 4, 1–33 Advance Access publication October 18, 2005

doi:10.1093/SER/mwj029

Social risk management through transitional labour markets Gu¨nther Schmid Director of the ‘Labour Market Policy and Employment’ Research Unit, Social Science Research Center Berlin (WZB) and Professor of Political Economy, Free University of Berlin, Berlin, Germany Correspondence: Gu¨nther Schmid, Wissenschaftszentrum Berlin, Reichpietschufer 50, D-10785 Berlin, Germany. E-mail: [email protected]

This essay takes the intrusion of the term ‘risk management’ into the social policy discourse as a ‘moral opportunity’ to reconsider the balance between solidarity and individual responsibility. The argument is developed in three stages: first, the psychology of intuitive beliefs and choices provides evidence of the bounded rationality of risky choices. The paper demonstrates how the institutionalization of opportunity structures recommended by the concept of transitional labour markets can overcome various kinds of asymmetries in risk perception. Second, imperfect or strategic information may also cause biased risk perception. Examples, therefore, are provided of how the analysis of labour market transitions throughout the life course can considerably improve the methodology of risk management. Third, the consideration of risk asymmetries and the reduction of information deficits do not yet provide any starting point for constructing acceptable risk sharing institutions. Normative principles of justice therefore have to be reconsidered. It is suggested that Rawls’ theory of justice should be enriched by Dworkin’s ethical theory and Sen’s capability approach. Keywords: risk management, rational choice, prospect theory, justice theory, labour market, social policy JEL classification: A130 relation of economics to social value, J400 particular labour markets, general, J650 unemployment insurance, J690 other (transitional labour markets)

1. Introduction ‘Once a word has entered the public domain, there it will remain for the next nine years’. (Zen Master Huang Long) Risk management has entered the public domain, whether we like it or not. ‘Risk’ apparently sounds less disturbing than poverty, sickness or unemployment. ß The Author 2005. Published by Oxford University Press and the Society for the Advancement of Socio-Economics. All rights reserved. For Permissions, please email: [email protected]

2

G. Schmid

If taken under control by good ‘management’, social risks look no more threatening than those involved in selecting a good financial investment. What has happened, many ask, to the ‘social’ element, to a responsibility that reflected concern and a community of interests? Have we really said farewell to the social element by adopting the inflationary vocabulary of ‘managing by this’ or ‘managing by that’? The answer is no. This essay takes the intrusion of risk management as a ‘moral opportunity’ to reconsider where the new balance between solidarity and individual responsibility might lie. The argument will be developed in three stages. First, it will be shown how risk perception has changed over the centuries, from resigned acceptance of one’s fate to a desire to master risk. The psychology of intuitive beliefs and choices, however, points to the bounded rationality of such choices. It will be demonstrated how the institutionalization of opportunity structures recommended by the concept of transitional labour markets (TLMs) can overcome various kinds of asymmetries in risk perception. TLMs seek to combine flexibility and security by establishing institutionalized ‘bridges’ at critical junctures in individuals’ working lives: as they move between education and work, between dependent employment and self-employment, between unemployment and employment, between paid work and unpaid work or incapacity for work and retirement. New forms of social security provision will cover not only the risk of involuntary unemployment but also the risks of other work-life discontinuities in order to encourage both men and women to engage in risky family, employment or human capital decisions during their life course, thereby making the labour market more flexible and open for groups at high risk of social exclusion.1 Second, imperfect or strategic information may also lead to biased risk perception. Proper risk analysis, therefore, must itself be a central element of social risk management. It will be demonstrated—using data from the European Household Panel and the European Labour Force Survey and with the aid of examples related to the European Employment Strategy and Member States of the European Union—how the analysis of labour market transitions throughout the life course recommended by the concept of TLM can considerably improve the methodology of risk management. Third, the consideration of risk asymmetries and the reduction of information deficits do not yet provide any starting points for constructing acceptable risk

1 The term transitional labour markets (TLMs) was coined for the first time by Schmid (1995). The concept was then developed by a network of researchers in Europe sponsored by the fourth and fifth (TLM.NET) framework research programme of the European Commission under the acronyms TRANSLAM and TLM.NET. The theoretical background, illustrated by applications and good practices, is best represented in Schmid (2002), Schmid and Gazier (2002) and Gazier (2003); for further publications and current research, the reader is advised to visit the website: www.siswo.uva.nl/tlm

Social risk management through TLMs

3

sharing institutions. Normative principles of justice have therefore to be reconsidered. It is argued that John Rawls’ theory of justice should be extended by Ronald Dworkin’s ethical theory and enriched by Amartya Sen’s capability approach, since it responds directly to the hidden agenda of risk management, namely the question of the conditions under which individuals could take on more responsibility. One of the central conditions is embedded employability, which is also a core element in TLM theory.

2. On the meaning and perception of risk If we look back in history, we can find substantive changes in the connotations of risk. In fact, as Peter Bernstein (1996) has shown in his remarkable story of risk, people’s attitudes have moved from resigned acceptance of one’s fate to a desire to master risk. On the other hand, recent social behavioural studies have increasingly drawn attention to the fact that people’s attitude towards risk is often biased by asymmetrical risk perception, which in turn leads to decisions that violate the utility maximization and inconsistency assumptions of the rational choice theory. Thus, the entry of the term ‘risk management’ into the public domain requires a brief look back at the etymology and perception of risks. 2.1 The changing meaning of risk The etymology of the word ‘risk’ can be traced back to the maritime traders of the North Italian city states of the 14th century.2 These traders were the first to insure against the pecuniary risks of voyages (Sinn, 1995), which were always something of an adventure. Rocky reefs and other hidden dangers had to be circumnavigated, and this is what the Latin word ‘risicare’ means. Winds had to be favourable, and this is what the French word ‘aventure’ hints at, since it means to sail ‘a vent’, that is before the wind. The wind not only propels ships but also brings news of unknown worlds and of new opportunities and challenges, which cosmopolitan people like knights, troubadours and sea captains were eager to capture. However, most of these people were not speculative risk-takers but calculating adventurers. Modern behavioural research has proven that the readiness to accept risk is highly correlated with the behavioural trait of control, caution and planning. From this perspective the risk management discourse actually opens up the prospect of calculated risk behaviour. It is based on the perception that risk is not only related to danger but also brings with it opportunities and gains. Thus, to risk something is a conscious choice and not a fate. 2

Cf. the enjoyable and profound little essay by the literary scholar Elisabeth Keller (2004) on the linguistic and cultural history of the concept of risk.

4

G. Schmid

According to German social philosopher Niklas Luhmann (1990), modern societies are characterized by the conversion of danger into risk. Luhmann argues that we can speak of danger only if the environment, natural catastrophes or evil adversaries cause the loss or damage. On the other hand, the word risk should always be used when the loss or damage can be attributed to an individual’s choice, including negligence. This shift in linguistic usage from danger to risk also gives the notion of modernization a more precise meaning. From this point of view, modernization today means to characterize loss or damage presumably caused by external factors as, in part at least, self-induced phenomena. Such phenomena require individual or collective steps to be taken in order to prevent, reduce or compensate any damage caused. Environmental damage is an intuitively plausible example of the shift from danger to risk. However, the same also applies to such apparently completely external risks as damage caused by earthquakes. The reason is that even damage of this kind can be mitigated, at least in part, by the appropriate construction of buildings or just by the avoidance of areas of highest risk or, as shown most recently and dramatically by the Southeast Asian tsunami, by establishing early warning systems. Similarly and with respect to labour market risks, dangers arising out of technological change or globalization are increasingly being converted into risks. It is fair to say that much can be done by employers and employees to prevent or to mitigate these external risks. It would, therefore, be mistaken to conclude that the change in linguistic usage simply reflects a process of individualization. Redefining danger as a social risk means nothing more, but also nothing less, than challenging the (old) established division of labour in risk management between individuals, families, firms and the state. The result of such a challenge may be that individuals bear, could bear or even should bear more risk. However, the outcome may also be that the emergence of new risks requires new forms of solidarity or collective insurance systems. For this reason, risk management should be regarded as a ‘moral opportunity’ to reflect on and to extend our knowledge of what we owe other members of our community (Heimer, 2003). Such reflections, finally, lead to ‘social risk management’—a term that has the advantage of re-emphasizing risk prevention and risk mitigation as alternatives to reactive risk coping (Holzmann and Jorgensen, 2000): On the supply side of the labour market, lifelong learning is obviously an example of a preventive approach to the risk of dismissal owing to skill obsolescence; an example on the demand side would be product innovations against the risk of wage competition through globalization or the risk of changes in consumer preferences. A key element in prevention is also ‘providence’, that is risk assessment based on professional forecasting models.


5

As far as risk mitigation is concerned, the acquisition of multiple skills or of general rather than firm-specific skills or the establishment of flexibility buffers through working-time accounts would be examples on the supply side. Built-in income stabilizers in the form of generous wage replacement benefits (at least in the short term) or the regional redistribution of purchasing power by means of universal insurance systems are examples of risk mitigation on the demand side. Risk coping: on the supply side, for example, solidarity requires risk sharing through wage replacement benefits on the basis of unemployment insurance or active placement in a new employment relationship; on the demand side, state expenditure programmes or the reduction of prime rate in order to kick-start the economy are needed. The examples listed for the three risk management strategies make it clear that the change of terminology from labour market policy to the social management of labour market risks (the same would also apply to social policy) at least has the advantage that the entire range of possibilities for action are considered. This avoids the danger of considering the dynamics of risks from the blinkered perspective of established policy areas or policy strategies. However, the optimistic view that risk can be managed rationally if only there is a will to do so is not justified. The psychology of intuitive beliefs and choice draws attention to the fact that a risky choice is not as rational as the rational choice theory would suggest. It is strongly influenced by the accessibility of information and the way in which risks are ‘framed’. 2.2 Asymmetries in the perception of risks Why do people perceive risks in a way that may lead to non-rational choices? The theory of intuitive judgements and choice (Kahnemann and Tversky, 2000; Kahnemann, 2003) highlights two mechanisms that cause asymmetric perception of risks: prospect evaluation and framing. Prospect theory contains three core ideas. First, the carriers of utility are events or transitions from one state to the other rather than states, and utility is assigned to gains and losses relative to a reference point, which is often status quo. Second, in assessing the prospects of events, losses usually loom intuitively larger than corresponding gains, and many experiments suggest that the loss aversion coefficient is about 2:1. Third, there is an endowment effect, whereby present or past experiences are intuitively given higher values than future prospects. The German comic talent Karl Valentin expressed this insight once in the statement: ‘In the past, even the future was better’. Thus, the transition from status quo to a gain of some sort is not valued in the same way as the transition from status quo to the loss or relinquishment of something.

6

G. Schmid

This often results in myopic decisions and violations of a substantive condition of rationality, namely utility maximization. Framing leads to the violation of another assumption of rational choice theory—the consistency of decisions. Framing theory refers to the fact that alternative formulations of the same situation make different aspects accessible to actual perception, thereby leading to different reactions. The same objective outcomes can be evaluated as gains or losses, depending on the framing of the reference state. For example, experiments have been conducted with a game known as the prisoner’s dilemma, once formulated as a Wall Street game and at another time as a community game. It turned out that participants were more inclined to cooperate in the community game than in the Wall Street game (reported in Shiller, 2003, p. 93). Another standard example is the ‘Asian disease’, in which people are confronted with two different descriptions of the same problem. If the reference state is framed as ‘all die’, people respond in a risk-averse way. Asked to choose between a health programme (A) that will certainly save one-third of the affected population (say 200 out of 600) and a programme (B) in which there is a 33% probability that all the affected population can be saved (but also a 66% probability that all will die), the majority (72%) choose the certain alternative A. If the reference state is framed as ‘no one dies’, people respond in a risk seeking way. Asked to choose between a health programme (C) that will lead to two-thirds dying for sure (say 400 out of 600) and a programme (D) that promises a 33% probability of nobody dying (leaving a two-thirds probability that all will die), the majority (78%) choose the risky alternative D. This rather robust inconsistency of choices particularly contradicts the invariance principle of rational choice. Loss aversion and variations in risk perception depending on framing lead to three hypotheses as crucial starting points for social risk management: First, if there is a choice between certain and uncertain gains, most people tend to be risk-averse. Since they are generally loss averse, they choose the certain alternative, even if the objective value of the prospective gain is (possibly much) greater than the value of the certain gain, which they would forfeit in choosing the uncertain alternative. In other words, they prefer the bird in the hand to the two in the bush. Second, if people have to make a choice between certain and uncertain losses, however, they tend to be speculative risk-takers. In this case again, their general loss aversion explains why they prefer the uncertain to the certain alternative, even if the objective value of the prospective loss is (possibly much) higher than the immediate and certain loss. In other words, they tend to behave like a player in the roulette game who tries to recoup his loss again and again until he/she ends up with nothing in hand.


7

Third, most people overestimate minor risks that might impinge on the present or very near future, such as the possibility of falling ill shortly before a planned journey. On the other hand, most people underestimate major risks that lie in the somewhat more distant future, such as becoming unable to work through disability or an enforced change of occupation. As a consequence, many people tend to insure themselves against possible disruption of their travel plans but not against the possibility of long-term disability; they are also less willing to save for any education or training that may be necessary in the future. A recent study of European attitudes towards the welfare state, for example, found that the majority of voters prefer under-insuring against unemployment (Boeri et al., 2001). These three insights derived from prospect and framing theory provide some useful hints for designing and gaining acceptance for a strategy of social risk management in the labour market: The appropriate strategy against risk aversion would be to provide a large opportunity set at critical points during the life course when risky decisions on transitions from one employment status to the other have to be made. Such an opportunity set would reduce the higher subjective valuation of the smaller but more imminent loss compared with the greater but uncertain gain by extending and securing the alternatives that would be available in the event of the first, risky choice failing, thereby increasing the probability that the greater gain might be realized. The preference for ‘one bird in the hand’ over ‘two in the bush’ might change if more bushes were in sight. Self-confidence might also increase if one such risky choice were to succeed. Establishing such a variable opportunity structure with ‘stepping stones’ or ‘bridges’ is one of the main objectives of the TLM concept (Schmid, 2002; Schmid and Gazier, 2002; Gazier, 2003). The appropriate strategy against speculative risk-taking would be to establish disincentives for gambling, for example, through high taxation of speculative gains and large-scale inheritance or of gains from winner-takes-all situations (Frank and Cook, 1995). From this perspective, the rationale of employment protection regulation can be interpreted as the avoidance of speculative search behaviour on the supply side and irresponsible hire and fire policy on the demand side, both leading possibly to high turnover costs for the whole economy. Activation policies would avoid the exploitation of unemployment benefits through moral hazards (Barbier, 2004; Schmid, 2003). The appropriate strategy against the overestimation of short-term (small) risks and underestimation of long-term (high) risks would be, on the one hand, to extend the expectations horizon for people engaging in risky employment relationships through social rights and entitlements, for instance to continuous

8

G. Schmid

training opportunities, and on the other hand, to establish obligations for preventing, mitigating or coping with major risks. For instance, incentives could be put in place to encourage preventive measures (such as training investment) or to ensure the compulsory participation in universal private or public insurance schemes (such as invalidity insurance). Some further examples may show how powerful the prospect evaluation and risk framing perspective is. On the basis of the tendency to overestimate shortterm small risks and underestimate long-term major risks, it can be reasonably assumed that individuals perceive the risk of being stuck in the low-wage sector to be greater than the risk of long-term unemployment, which may result from being too choosy about the jobs they will accept. Active labour market policies, therefore, should not be confined solely to offering jobs and placing individuals in work. Follow-up measures are required for transforming mere workfare measures into stepping stones for a sustainable career in the labour market. Conversely, the barriers to the acceptance of a risky job should not be underestimated either. Such barriers may arise out of the fact that the acceptance of risky jobs means the abandoning of familiar certainties, even though they may have a lower value than the new employment prospects. These ‘familiar certainties’ may be of various kinds. The reliability of social assistance benefits possibly supplemented by some small clandestine employment may be one example; the confidence in one’s own productive capacities, another. Taking on a risky new job, however, brings with it the fear of losing these capacities. Daniel Bernoulli, one of the founders of the probability theory and thus of risk management, provides a nice example: a beggar will not give up begging for a workfare job since he would lose his ability to beg. He has to be offered something more.3 The consequence of this insight again is to overcome risk aversion by offering a broad opportunity set of transitions throughout the life course, such as (a) the opportunity to try out several jobs without the immediate withdrawal of benefits if one option does not lead to success at once—a rule that excludes rigid workfare strategies which do not allow trial and error as a productive job search strategy; (b) the possibility of maintaining entitlement to unemployment benefits if an individual takes on the risk of self-employment, for example; (c) the possibility of wage insurance, if a person accepts the risk of a low-paid job; (d) the possibility

3 Bernoulli says that tangible assets and financial claims are less valuable than productive capacity, including even the beggar’s talent. He suggests that a man who can earn 10 ducats a year by begging will probably reject an offer of 50 ducats to refrain from begging, because after spending the 50 ducats he would have no way of supporting himself. There must, however, be some amount that he would accept in return for a promise never to beg again. If the amount were, for instance, 100 ducats, ‘we might say that [the beggar] is possessed of wealth worth one hundred’. (Bernstein, 1996, p.119).


9

of building up entitlement to benefits again under carefully selected circumstances; this entitlement effect of insurance is to some extent also effective against illegal work; (e) the granting of tax or social security credits on acceptance of a risky new regular job; (f) the possibility of in-work benefits under certain circumstances, as in the case of family or other care obligations.4 Thus, social risk management means that risks should be framed appropriately and policies designed in such a way so as to reduce or compensate asymmetrical risk perceptions. In order to realize both objectives, more incisive risk analysis is required in order to eliminate or at least mitigate the informational asymmetries underlying non-rational risk perceptions. Thus a discussion of the appropriate risk analysis is called for.

3. On the methodology of risk analysis What are the old and what are the new risks? Here, the TLM theory draws attention to the increasing importance of internal or ‘manufactured’ risks compared to the external risks that predominated in the past (Beck, 1992; Giddens, 1998, 2003). How can risk assessment be improved to support preventive strategies intended to achieve the 3-fold objective of preventing, mitigating and coping with risks? Here, the TLM theory suggests that risks should be analysed in terms of transitions from one employment status to another (or of sequences of such transitions) in order to assess the consequences of such mobility for earnings, job quality (including career prospects) and social security. 3.1 The new labour market risks: some stylized facts Three stylized new risks5 are relatively well documented in labour market research: (a) the increasing risks of social exclusion related to failures in education and training; (b) the increasing risks of precariousness related to temporary jobs and compressed work careers; and (c) the increasing risks of long-term unemployment or forced inactivity related to the erosion of internal labour markets. (a) If we take the main indicator of full employment in the European Employment Strategy, namely the goal of reaching an employment rate of 70% by 2010, then a breakdown by qualification immediately shows where the main problem lies (Figure 1): the employment rate of high-skill workers surpasses the 4

Examples of good practice from a TLM point of view drawn from European countries can be found especially in Schmid and Gazier (2002), Gazier (2003) and, with particular reference to Germany, Schmid (2002, 2003). 5

For a complementary approach to analysing new risks and welfare regime adjustments cf. Sarfati and Bonoli (2002) and Taylor-Gooby (2004).

10

G. Schmid

Figure 1 Employment rates by age (25–64) and skill level in percentage (2002). Source: OECD Employment Outlook 2004, Table D.

benchmark by about 15 percentage points, and this holds true irrespective of employment or welfare regime. It is low-skill workers whose opportunities for participating in the labour market are seriously damaged. It is important to look also at the other side of the coin, which is unfortunately not well reflected in the European Employment Strategy, namely unemployment rates (Figure 2). If the Lisbon Strategy had included as one of its benchmarks the halving of the unemployment rate to the level of 5% by 2010, it is clear that, with a few exceptions, high-skill workers are already at this level or even below, whereas in many countries the 5% benchmark is desperately out of reach for low-skill groups. (b) The second concern is the increasing amount of temporary work, either in the form of fixed-term contracts, temp-agency work or contract work, which is often disguised as a form of self-employment. Figure 3 shows the level of fixedterm contracts in 1985 (delta characters) and 2002 (bars). With the exception of Denmark and the United Kingdom, fixed-term contracts increased in some countries drastically. The correlation of this increase with the OECD Index of dismissal protection is quite strong (r ¼ 0.44), suggesting a reaction to institutional rigidities. This assumption corresponds well with the two exceptions, Denmark and the United Kingdom, where dismissal protection is almost unknown. Not so well known, however, is the concentration of these precarious forms of work among the young. The case of Germany is especially striking. The burden of risks related to fixed-term contracts falls almost completely on young people


11

Figure 2 Unemployment rates by age (25–64) and skill level in percentage (2002). Source: OECD Employment Outlook 2004, Table D.

Figure 3 Employees on fixed-term contracts as a share of all employees (15–64), 1985–2003. Source: Eurostat New Cronos Data Bank.

aged between 15 and 25 and on young adults aged between 25 and 35. Many studies have shown that fixed-term contracts are often useful bridges to regular work. However, for many young people (and in some countries even for the majority), fixed-term contracts are unfortunately also traps that lock them into

12

G. Schmid

Figure 4 Dependent employees by fixed-term contract and age 1984 to 2002 (excluding trainees). Source: German Federal Statistical Office, Technical series 1, Population and Employment. Series 4.1.1, State and evolution of employment (Microcensus results).

permanently disrupted job careers and often lead, ultimately, to social exclusion (Figure 4).6 These risks are often aggravated by ‘compressed work careers’. This new risk arises out of the need to fulfil several social roles simultaneously within a short period of their working life. It mainly affects young women aged between 20 and 35. Since labour market participation is becoming the norm for these women, they have to come to grips with at least five social roles virtually simultaneously. They have to acquire a good education, look for a suitable job, plan a sustainable career, select a suitable partner and set up a family at considerable expense for housing and furnishing. The way work, education and welfare (including the housing market) are organized today scarcely help them to cope with these diverse roles. For this reason, their transition into a sustainable employment career is seriously endangered. Even if they are successful, the associated pressures can lead to physical or psychological disturbances. A study carried out in The Netherlands by de Bruijn et al. (2003) has revealed a dramatic increase in the incapacity to work among young women, while an Australian research team even speaks of the ‘excluded generation’ (Macdonald and Holm, 2001). The relative neglect of this problem for young adults compared to the attention given to the elderly is a serious defect in the European Employment Strategy. (c) This does not mean, of course, that efforts should not longer be made to deal with the increasing risks for older workers, as reflected both in employment rates below the full employment benchmark for most EU member states

6

See, for instance, Scho¨mann et al. (1998) and Mangan (2000).


13

Figure 5 Employment rates of older workers (55–64) in percentage, 1983–2003. Source: OECD Employment Outlook 1997, 2004: Table C.

(Figure 5) and in the high risk of long-term unemployment. These risks have to be considered against the background of the ‘erosion of internal labour markets’ or the ‘crisis of the open-ended employment relationship’, recently well explained by Gautie´ (2004), Marsden (2003), Supiot (2001) and others. From the perspective of risk management, the backbone of internal labour markets is an implicit insurance contract: the employer offers to the risk-averse male breadwinner job security and earning stability over his life course in exchange for the acceptance of wages below the productivity level achieved at the height of his work career. This implicit insurance contract is breaking down without clear alternatives in sight. What could these alternatives be? This question leads to the second point on the methodology of risk analysis: How can risk assessment contribute to the formulation of suitable responses to the new labour market risks? 3.2 On the measurement and presentation of risks In his highly readable book Reckoning with Risk, Learning to Live with Uncertainty, Gigerenzer (2002) recommends that, whenever possible, frequencies rather than probabilities should be used in presenting risks. Frequencies are intuitively better understood, whereas probabilities are often disturbing. Gigerenzer cites a good example of the weather forecast: ‘There is a 30% probability that it will rain tomorrow’ is a probability statement about a single event. Confronted

14

G. Schmid

with this statement, people draw quite different conclusions. Some think it will rain 30% of the time, others that it will rain in 30% of the area, and a third group believes it will rain in 30% of the days that are like tomorrow. On the other hand, ‘It will rain 10 days in May’ is a statement about frequency. This may be either true or false and is thus not open to ambiguity. A single probable event, however, can never be true or false unless the probability is 1 or 0. For the decision to invest in an umbrella, the frequency statement is—given the level of risk aversion— more helpful than the probability statement. It is reasonable to think that the same holds true for investment decisions in unemployment insurance or labour market policies. In labour market research, however, we often find probability statements, which tend to be very opaque. Good practice of how risk assessment could be improved by the use of frequency statements will therefore be presented. The example is taken from the third chapter of the Employment in Europe Report 2002, which explicitly adopted the TLM framework (European Commission, 2002). Some stylized facts on frequencies of labour market transitions may be useful at the beginning. Almost 50% of the workers who became unemployed in 1997 had found new employment in 1998. However, less than one-fourth of those who were unemployed in 1996 and 1997 had found new jobs in 1998. Less than one-fifth of the long-term unemployed in 1996 had found new jobs in 1998. Information of this kind is sufficiently alarming to move policy in the right direction: everything needs to be done to prevent long-term unemployment from the outset. In order to find the proper design for policy intervention, however, such frequencies have to be combined with policies. For instance, in 1997/98, 4.6% of all employees in jobs without training opportunities became unemployed; the risk for low-skill workers in such jobs was significantly greater (5.5%) than that for high-skill workers (3%). On the other hand, in the same years, 1.6% of all employees in jobs with training opportunities became unemployed; the risk for low-skill workers in such jobs was no greater (1.86%) than that for high-skill workers (1.93%). It follows from these figures that even poorly qualified workers should be given opportunities to receive training in the workplace in order to reduce their risk of unemployment. Additional information on the outcome of such policy interventions enriches the policy-relevant risk assessment further. For instance: (a) the annual productivity increase (1995–2001) correlates strongly with the rate of participation in firm-based further training; (b) a 1% increase in participation in further training in Germany leads to a 0.3% increase in productivity; and (c) objective job quality correlates strongly with self-reported job satisfaction, which triggers further productivity increases. One important strategic conclusion can be drawn from these pieces of information taken as a whole, which is that the reintegration of the long-term


15

Table 1 Transition matrix for employment policy targets (EU14) Status at time t (1998) Status at time t (t 1)

Inactive

Unemployed

Low-skill job

High-skill job

Inactive Unemployed

87.5 17.5

5.0 52.5

5.0 20.0

2.5 10.0

Low-skill job

7.5

12.5

50.0

30.0

High-skill job

2.5

2.5

5.0

90.0

Source: Employment in Europe 2004.

unemployed into the labour market on the basis of a series of stepping stones can be effective. In other words, work first plus various options for accumulating work experience and training is a sensible employment strategy. Finally, the effects of such strategies can be simulated by means of transition matrices. The following transition matrix (Table 1) shows realized transitions between 1997 and 1998 in 14 EU member states. What does the matrix tell us? In 1997, 87.5% of the population fit for work but economically inactive were still inactive in 1998; 5% of the inactive population moved into unemployment, 5% into poor-quality jobs and 2.5% into good jobs. Of the unemployed population, 17.5% moved into activity, more than half remained unemployed, only 20% went into poor-quality jobs and 10% into good jobs. Of the low-skill people, 7.5% became inactive, 12.5% unemployed, 50% remained in unskilled jobs and at least 30% moved up. The overwhelming majority of workers in good jobs (90%) were still in good jobs in the following year. Some of these transitions are very disappointing but may be explained to a large extent by the poor economic situation in the base year 1997. In order to be able to make predictions over a longer period, it would certainly be more sensible to take as a starting scenario the average values from several years, possibly even of an entire business cycle. For a thought experiment, however, we can use this matrix as a baseline for estimating the impact of employment strategies aimed at speeding up ‘good’ transitions and preventing or reducing ‘bad’ transitions. A simulation of these effects carried out by the European Commission produced the following results: (a) The increase in the rate of transition from bad to good jobs from 30 to 40% would increase the employment rate over 10 years by one percentage point; (b) the reduction in the rate of transition from bad jobs into unemployment from 12.5 to 7.5%, with a simultaneous provision of training opportunities, would increase the employment rate by a further 1.5 percentage points; (c) the increase in the rate of transition from unemployment into bad jobs from 20 to 25%, again combined with training opportunities, would increase

16

G. Schmid

the employment rate by a further percentage point; (d) if all the measures were combined, then the employment rate would increase by almost four percentage points, while the unemployment rate would be reduced by two percentage points. In recent years, considerable progress has been made in collecting information on risks related to transitions. In particular, the Employment in Europe Report 2004 (Chapter 4) contains a rich set of this type of information, both in descriptive as well as in econometric terms. Two additional transition matrixes from this analysis may give a taste of this progress. For example, multi-year transitions by main economic activity for EU15 show that only 0.5% of permanent employees moved into education or training in the 6 year period 1995–2001. This low figure might be a concern given the need for greater adaptability in the low-skill segment of the adult workforce. Of temporary employees, 55% moved into permanent employment; however, one in five moved into inactivity or unemployment. Stability among the self-employed was almost as high among the permanently employed, and almost one-eighth actually moved into permanent employment (Table 2). The multi-year transition matrix by pay level shows, for example, that only 26.2% of low-paid workers were still low-paid after 7 years; 44% (39.2 plus 4.8) moved to the medium or high pay brackets. On the other hand, however, almost 30% moved to no pay, which means they became unemployed or inactive or disappeared into the informal or even illegal sector (Table 3). The commission has developed an interesting mobility index based on the aggregation of several transitions between employment statuses in the period from 1994 to 2002. This index correlates strongly with the employment rate (see Figure 6). The causal direction, of course, is unclear, but it seems to be a plausible hypothesis that a sustainable increase in the employment rate requires higher mobility rates. Table 2 Multi-year transitions by main economic activity (EU15 total, row percentages) tþ6

Temporary

Selfemployed

Not employed

Education/ training

t ¼ 1995

Permanent

Permanent

76.7

4.0

3.1

15.8

Temporary

55.0

16.4

6.4

20.7

1.5

Self-employed

12.5

2.9

68.3

15.8

0.5

0.5

Not employed

18.7

6.1

4.8

69.0

1.3

Education/training

41.6

13.4

4.8

16.1

24.1

Source: Employment in Europe 2004, Table 50.


17

Table 3 Multi-year transitions by pay level (EU15, row percentages) tþ7 t ¼ 1995

No pay

Low pay

Medium pay

High pay

No pay Low pay

62.4 29.8

9.1 26.2

22.7 39.2

5.8 4.8

Medium pay

17.4

5.2

61.3

16.1

High pay

17.0

0.9

17.0

65.1

Source: Employment in Europe 2004, Table 54.

Figure 6 Mobility index and employment rate. Source: Employment in Europe 2004; own calculations.

This assumption is supported by the fact that mobility is positively correlated (r ¼ 0.44) both with transitions from inactivity to employment (indicating an open labour market) and (r ¼ 0.55) with transitions from temporary to permanent jobs (indicating a kind of job-to-job security). However, as Figure 7 shows, higher mobility seems also to be related to fewer transitions from low pay to higher pay (r ¼ 0.46). Again, the causal relationship is unclear, but it is reasonable to be concerned about the possible negative impact of employment status mobility. A possible interpretation could be that modern labour

18

G. Schmid

Figure 7 Mobility index and transitions from low pay to higher pay. Source: Employment in Europe 2004; own calculations.

markets—exposed increasingly to international competition—increasingly require ‘active’ social spending that attempts to change the distribution of market income by promoting labour market participation among the segments of the population with lower than normal market income. Such an active spending policy would also make higher mobility more acceptable if it linked job mobility with upward status mobility or at least with retention of the same income status. Acceptance of mobility, however, requires workers to feel they are being treated fairly. There is increasing empirical evidence that social expenditure as such is not detrimental to adaptability. Far from there being any intrinsic contradiction between an efficient and dynamic economy and one that places social justice at its core, the achievement of the former seems to require the latter, albeit in a form that directs redistribution into investment in the individual capacity to participate in the modern economy (Arjona et al., 2001; Schmid, 2003). Of course, a more sophisticated analysis is needed to endorse such informed but still speculative interpretations.7 Moreover, from the fact, as shown above,

7 The Employment in Europe Report 2004 already contains some interesting Logit-Models for transitions from ‘non-working’ to ‘working’ and for transitions from ‘lower pay’ to ‘higher pay’. For good practice of risk analysis in the vein of TLM, see O’Reilly et al. (2000), de Koning and Mosley (2001), Scho¨mann and O’Connell (2002), Muffels et al. (2002), Neugart and Scho¨mann (2002), Meager and Bates (2002) and Gangl (2003); with special emphasis on the gender dimension cf. Mosley et al. (2002).


19

that ‘manufactured’ risks are increasing and that preferences with respect to employment status8 change during the life course, it is reasonable to conclude that the policy and politics of risk sharing between individuals and society has become more complicated than ever if they are to meet the criteria of social justice. Whereas the theory of intuitive beliefs and choice supported the conclusion that equity-sensitive opportunity sets might help to overcome myopic choices in labour market transitions, positive risk analysis has produced evidence that a kind of job-to-job security or stepping-stone strategy might combine flexibility with security. However, even the best risk assessment does not answer the nagging question of who is to be responsible for preventing, mitigating or coping with the new risks in the labour market. It is time, therefore, to tackle the ‘moral opportunity’ related to the shifting terminology from social policy to risk management from a normative point of view as well. What might be the criteria for a new balance of public–private risk sharing?

4. Normative principles of risk sharing The two main schools reflecting such criteria are welfare theory and justice theory. Put simply, the welfare theory is based on the principle of utility maximization. Apart from the guarantee of a minimum level of subsistence for all, a society is regarded as just if it maximizes utility for all or, as the highly influential philosopher Jeremy Bentham put it, if it produces the greatest happiness of the greatest number. Any income distribution is fair, provided it increases average utility and leaves no one individual any worse off. The Justice theory, on the other hand, which can be traced back to Kant, holds that individual utility cannot be compared and that it is therefore non-sensical to talk of maximizing average utility. Furthermore, an unequal income distribution can be justified only on the basis of the generally applicable and mutually agreed criteria. Since the naı¨ve utility maximization argument can no longer be sustained in the wake of Arrow’s (1974) proof of the impossibility of the consistent aggregation of individual utilities, the following discussion focuses on the justice theory. Currently, the two main schools are those of John Rawls and Ronald Dworkin. 4.1 Contractarian or ethical justice theory? John Rawls versus Ronald Dworkin John Rawls (1990, 2001) arrives at the generally agreed criteria for justice from his starting point of the familiar theory of the social contract. Behind what he terms the ‘veil of ignorance’, most people would decide in favour of the so-called difference principle: inequality is justified, provided it improves the position of the 8

For example with respect to full-time or part-time work, dependent employment or selfemployment, combining work and education, work and family, work and part-time retirement.

20

G. Schmid

economically most disadvantaged group. Put simply, Rawls argues for the maximization of the minimum (the maximin principle), according to which alternatives are ranked by their worst possible outcome. The alternative to be adopted is the one whose worst outcome is better than the worst outcomes of the others. Welfare theorists or utilitarians, on the other hand, argue for the maximization of the average. The differing consequences of these two strategies (maximizing the minimum and maximizing the average) can be demonstrated by means of a simple calculation. Let the initial average distribution be 1000 value units and let the lowest quintile be 200 and the highest quintile 2000, giving a ratio of 1:10. Let us assume that welfare and distribution rise on average to 1500 value units, with the lowest quintile again being 200 and the highest quintile 3000, which gives a ratio of 1:15. This distribution would be acceptable to welfare theorists but not to Rawls. If, on the other hand, we assume that the average rises to only 1200, with the lowest quintile being 300 and the highest 2000, which gives a ratio of 1:7, this distribution would be acceptable to Rawls but not to welfare theorists. Ronald Dworkin (2000) joins Rawls in his criticism of the utilitarian theory of justice. However, he sees severe problems in applying Rawls’ social criteria of redistribution. First, the difference principle offers no advice as to where the ceiling defining the worst-off class should be drawn. Second, it legitimizes any inequality as long as the situation of the worst-off improves even by only a tiny margin.9 Third, it does not appeal to the majority of people. It draws attention only to the position of those who have the fewest primary goods. Those who are better off but nevertheless have to struggle to secure a decent living for their families are neglected. The difference principle, therefore, seems to be most appropriate in times of rising general expectations. In times of dramatic social and economic change, however, it is highly probable that the middle classes will also be much affected. It is hardly surprising that these losers (or people at risk of losing) feel great resentment when part of their hard-earned wage is taken in taxes and paid over to those who do not work at all. Dworkin’s main objection, therefore, is that Rawls’ difference principle is ethically insensitive. It does not build on the basic distinction between the causal effects of (external) circumstances and individual choices. It neglects individual

9 This objection might be contested depending on how Rawls’ first principle of justice, equality in primary goods, is interpreted. If ‘having a decent job’ were subsumed under primary goods, then the requirement for redistribution would be much greater than in the opposite case. In his restatement, Rawls seems to give equal access to education the status of a ‘basic liberty’; he also acknowledges that the state has to ensure that ‘excessive concentration of property and wealth’ has to be prevented (Rawls, 2001, p.43). For an illuminating discussion of the theory of justice as it relates to welfare state reforms see also Standing (1999), Esping-Andersen et al. (2002) and Vandenbroucke (2001).


21

responsibility for outcomes under given circumstances such as differences in talent or differences in exposure to economic change. To be responsible for the consequences of individual choice, however, is a crucial and widely accepted ethical principle. It also underlies the notion of risk management, which assumes the active participation of individuals in responding to external challenges. What is required, therefore, is an ethically acceptable balance of individual rights and obligations. A fair distribution is one that takes into account both the circumstances that produce inequalities in living standards or quality of life and of the individual choices that affect these standards. From this point of view, individuals cannot be held responsible for distributions caused by external factors. Factors that individuals cannot influence include natural talents, handicaps and susceptibility to certain diseases, external phenomena such as weather and natural environment and, finally, bad luck or good fortune ensuing from the random conjunction of events. However, individuals must be made responsible for distributions resulting from their own choices, such as a personal decision to work less, have certain (exotic) preferences or choosing a particular occupation. Individuals can be made responsible for their decisions, however, only as long as they are endowed with the same resources. Dworkin means this very literally. Since income distributions are determined by chance and external circumstances, there must be periodic redistributions within and between generations, for example through high inheritance and progressive income taxes.10 4.2 Ideal types of social risk management Dworkin’s ethical principle of justice can well serve as a basis for the normative foundation of social risk management. Depending on whether the risks are triggered or caused by individual choice or external circumstances and whether the consequences of the risks can be borne individually or exceed individuals’ capacity and should therefore be managed collectively, four ideal types of social risk management can be identified and applied to the labour market; in reality, of course, the boundaries between these ideal types are blurred (Figure 8). Type 1—‘individual responsibility’—applies to cases in which the risks are predominantly the result of personal decisions and can largely be covered by private or self-organized collective insurance schemes. This type includes private savings policies and self-organized collective income maintenance schemes 10

The Yale economist Robert Shiller (2003) was thinking along the same lines when he recently proposed a general insurance scheme against inequality. He suggested that the tax system should be reorganized in such a way as to ensure that tax rates were geared to a democratically decided and politically determined distribution structure. A certain level of the Gini coefficient, for example, could be used as a benchmark.

22

G. Schmid

Actors’ risk management capability Individuals Risks resulting from

Individual choice External circumstances

(I)

Individual responsibility

(III) Individual solidarity

Society (II)

Solidarity

(IV) Collective solidarity

Figure 8 Ideal types of social risk sharing. Source: G. Schmid.

(in Germany, for example, collective agreements provide for top-up payments for short-time working benefits), implicit insurance agreements in the form of seniority pay and agreements on protection against rationalization, for example and company pacts providing, for example, employment guarantees in exchange for wage concessions. The notion of private responsibility can also be extended to include the employment contract as a form of insurance policy: in exchange for guaranteed wage payments and a greater or lesser degree of employment protection provided by the employer (the principal), the employee (agent) undertakes to carry out in future as yet unspecified work tasks, the content of which is to be determined by ad hoc instructions or by the worker’s own flexible reactions to unforeseen changes in the external environment or in requirements. Seen from this standpoint, intelligent workers who contribute to decisionmaking and accept joint responsibility are a key element in an effective system of employability insurance. Type II—‘solidarity’—concerns cases in which the risks are indeed triggered by individual decisions or predispositions, but the ability to make individual provision (cf. the risk prevention and reduction strategies) is low or the loss or damage incurred by the individual affected is so great that immediate collective or public assistance is required. To cite one drastic example: in a road traffic accident causing serious injuries, the emergency doctors do not ask who was to blame. When human life and dignity are at stake, the cause of whatever is threatening them is irrelevant to the decision to offer solidary assistance. There is a human right to solidarity. A labour market example would be an individual who chooses a risky occupation, such as a singer or a dancer and then suddenly loses his or her voice or the ability to jump because of an accident. In such a situation, we expect the guarantee of a basic income and the restoration of equality of opportunity through the provision of rehabilitation measures. Type III—‘individual solidarity’—concerns cases in which the risks are externally caused, that is they lie outside the sphere of influence of those affected, but their effects can be mitigated or even eliminated by individual or self-organized collective efforts. Natural catastrophes seldom occur in labour markets, but


23

company failures caused by spectacular mismanagement could be included under this heading. Otherwise, structural changes, adverse economic situations as well as political decisions, such as the expansion of the EU, can all trigger external labour market risks. Nevertheless, by adopting ‘activation strategies’, individuals, households, firms and associations can do much to prevent or alleviate external risks or deal with them once they have materialized. There is no need to list in detail such individual contributions to risk reduction. Besides intensive involvement in the search for a new job or participation in any training programmes that may be necessary, they also include, where necessary, the willingness to relocate. Individual learning and working-time accounts are relatively new measures that can help to protect against such risks through individual solidary efforts. Type IV—‘collective solidarity’—encompasses cases in which the risks have an external cause and their consequences cannot be dealt with by individuals acting alone or even in private collective associations. Since unemployment, for example, is ‘catching’, that is, it gives rise to further unemployment through chain effects (a shortfall in demand triggers other shortfalls) or even imitation (if company A breaks the social taboo on mass redundancies, then company B will follow suit), external effects can be adduced in further justification of public intervention and/or attempts to deal with the consequences. The established system of unemployment insurance and the public employment service are examples of collective responsibility. However, developments during the 1980s and 1990s, such as the abolishment of the monopoly for public employment service, the opportunities opened up by contract management or the new insurance technologies and the current debate on privatizing unemployment insurance11 are good reasons to call into question not the entire notion of state responsibility but the ways in which collective solidarity is currently organized. As the increasingly common notions of ‘governance’, the ‘ensuring’ or ‘enterprising state’ remind us, the extent to which solidarity, individual solidarity or even individual responsibility might represent functionally more efficient alternatives to the state as guarantor has to be repeatedly assessed.12 As the preceding reflections suggest, these four ideal types indicate some of the general rules that might be applied in choosing from among these four alternative models of allocating private and public responsibilities. The most general rule relevant to social risk management is that—apart from pure individual responsibility—individuals or self-organized collectivities (associations) can take over more responsibility only if they are empowered to do so. As far as the risk of unemployment is concerned, Dworkin

11 12

Cf., for example, Breyer et al. (2004), Feldstein (2005), Fo¨lster (1999) and Stiglitz and Yun (2002).

Cf. on this subject, among others, Considine (2001), Daly (2003), Giddens (2003), Heimer (2003) and Williamson (1996); in Germany among others Hoffmann-Riem (2001) and Schuppert (2003).

24

G. Schmid

himself proposes a mental game. If people in the ‘hypothetical insurance system’ were to be asked what insurance they would be prepared to purchase—provided resources and risks were equally distributed—then the majority would not argue in favour either of a generous or of a meagre unemployment insurance scheme. Presumably, most people would vote for a ‘compulsory contribution system’ that is relatively generous, but only on the condition that the insurers actively encourage the unemployed to take up new jobs (particularly by providing training) and the unemployed themselves accept reasonable jobs and training programmes. The result, therefore, is some mixture of public and private responsibilities and obligations. Dworkin’s argument sounds plausible but obviously requires further clarification. Otherwise, it amounts to little more than plain old-fashioned common sense or, in the language of welfare regimes, to the Panglossian assertion that a combination of the liberal, social-democratic and conservative welfare regimes could produce the ‘best of all possible worlds’.13 It remains unclear what endowment with the ‘same resources’ or acceptance of ‘reasonable jobs and training programmes’ means. 4.3 Equality of resources or equality of capabilities? From ‘freedom from want’ to ‘freedom to act’ Amartya Sen’s (2001) notion of capability can assist here. For Sen, material equality is at best a necessary but not sufficient condition for a fair distribution of life chances. What is more decisive is individuals’ ability to convert the resources available to them into the states they themselves aspire to, that is a flexible endowment with resources that may be quite unevenly distributed but which enables all individuals to realize the plans they have for their own lives. In Sen’s views, material freedoms play an important role in such a functional capability. As far as the labour market is concerned, this would mean the freedom of choice among different occupations and employment forms, the possibility of reconciling work and private life and opportunities to regenerate or extend capabilities throughout the life course. Sen’s capability approach remains rather abstract. However, with some imagination and modification, it could breathe new life into the notion of ‘employability’. Gazier (1999) persuasively described the evolution of this concept as a story of continuous enrichment, distinguishing seven stages: dichotomized employability, socio-medical employability, manpower policy employability, 13

In his novel Candide ou l’optimisme, Voltaire lampoons the notion, attributed to Gottfried Wilhelm Leibniz, that the existing world is also the ‘best of all worlds’, putting it into the mouth of Dr Pangloss, Candide’s tutor.


25

transitional employability, labour market performance employability, initiative employability and interactive employability. Sen’s capability concept applies to the last three notions. It suggests that the concept of employability could be fruitfully extended by the addition of three further dimensions: sustainability, autonomy and embeddedness.14 Let us look first at sustainability. To be of working age (currently 15–64), healthy and to have a basic education is no longer sufficient to have a successful working life. Not only is the required basic stock of education increasing all the time, but there is also a growing need constantly to renew or expand this stock over the life course. The conversion of employability into employment also requires the enhancement of transferable and general skills, which reduces both the risk of hold-up situations inherent in firm-specific investment and that of individual long-term unemployment in case of firm closure or bankruptcy. We turn now to autonomy. Sen’s approach reminds us of the possible restrictions of the capability for work over the life course. These may be physical restrictions but they may also arise out of social obligations, especially care for children, for the chronically ill and for other dependent people. These restrictions of work autonomy have to be compensated for. The gender dimension received insufficient attention in Sen’s analysis. In a special issue of the journal Feminist Economics, Nussbaum (2000, 2003) and others sought to remedy this neglect. In addition, Lewis and Guillari (2005) argued recently in favour of a list of basic capabilities to be evenly distributed. These capabilities go beyond material resources and include, for example, the capability of undertaking caring and parenting activities. As long as men do not make an equal contribution to such activities, women’s material freedom to choose an occupation and to use their occupational capacities to their full extent will remain restricted. The third and final dimension is embeddedness. Since individual choice and external circumstances are, like individual and collective capacities to act, increasingly intertwined, functioning networks are required in order to

14

Dichotomized employability is the simple distinction between able-bodied ‘employable’ persons with a minimum of general education and all other ‘unemployable’ persons excluded from the labour market. In socio-medical employability, degrees of physical and health capacities are taken into account in order to define employability as the distance between existing work abilities and specified work requirements. In manpower policy employability account is taken of people’s various disadvantages arising out of individual traits and labour market conditions; the ‘active labour market policy’ is supposed to close the gap between the potential and full employability of disadvantaged groups. Transitional employability draws attention to discontinuities of work during the life course as a possible cause of declining employability and to the resultant need to support the ability to move quickly in and out of employment, to make transitions between full-time and part-time jobs and to upgrade skills for entering new jobs. I have slightly modified the last three categories of Gazier’s typology.

26

G. Schmid

constantly renew social capital. It is not sufficient—as advocates of the ‘third way’, ‘shareholder capitalism’ or the ‘social investment state’ suggest15—to endow everybody with the necessary resources and then abandon them to market forces. Competition is a fantastic vehicle for increasing efficiency where markets are functioning. However, the conditions for properly functioning markets are seldom met. Even markets need formal regulatory infrastructures; even markets have to be embedded into trust relationships; and even markets are unfair in being often a lottery game and not a competitive game, resulting in outcomes for which individuals cannot be held responsible.16 Thus enhancing the capacity of individuals to shoulder more responsibility requires the concept of ‘employability’ to be extended by including additional dimensions of job quality: Sustainable employability, which enables individuals to prevent, mitigate or cope with revolving door effects and repeated disruption of employment careers. Workfare policies for low-skill workers need to be supplemented with additional career development measures, whereas workers already in possession of skills and qualifications need to have access to continuous training opportunities. Functioning employability can be defined as the ability of individuals to realize their own life plans in the labour market, to change these plans and to make transitions between various kinds of employment, for instance between parttime and full-time work or between dependent work and self-employment or a combination of the two. The ability to act, to be a relatively independent agent of one’s own fate, also requires communication skills and the ability to tolerate ambiguities in job descriptions as well as the risks of sunken investments related to highly specialized jobs. ‘Freedom from want’ is still a central aim of European social policy; however, ‘freedom to act’ is nowadays an equally highly esteemed value requiring an enriched concept of employability.17 Embedded employability18 goes even further and addresses the need to establish functioning opportunity structures to guarantee meaningful and decent employment over the life course. A functioning opportunity structure can be defined as the effective interplay between private, social and public partners mediated by proper institutions and laws. The idea of meaningful and decent work introduces both a social norm of minimum earnings, minimum security

15

See, for example, Giddens (1998) and Le Grand (2003).

16

See, among others, Solow (1990) and Schmid (1993).

17

In this vein see also Begg and Berghman (2002) and Salais (2003).

18

I have taken this term from a discussion with Anton Hemerijck.


27

and health conditions at the work place, as well as options for individual self-development and participation.19 Embedded employability also relates to functioning employability by emphasizing the public aspect of work in terms of recognition, reward, self-respect and mutual respect—in other words, all the non-instrumental elements of work (Hirschman, 1982). All three elements of an enriched concept of employability can be realized only if the legal context of the employment relationship is adapted to the new balance of individual responsibility and solidarity that is required. The emergence of new social rights, as described so forcefully in the Supiot Report, gives hope as well as direction for the development of a social citizenship that goes beyond employment protection or social security (Supiot, 2001, p. 228). These social rights are new in that they cover subjects unfamiliar to industrial wage-earners: right to training, to appropriate working hours, to a family life and to occupational redeployment or retraining. Their scope is also new since they would cover not only ‘regular’ wage-earners but also the self-employed, the semi-self-employed, and temp-agency, contract and marginal workers. Finally they are genuinely new in nature, since they often take the form of vouchers or social drawing rights, which allow workers to draw on collective solidarity, within defined and (possibly) collectively bargained limits in order to exercise the new freedoms. Therefore new forms of security develop, which are no longer seen as being given in exchange for subordination (as in the old employment contract) but as the foundations of a new freedom to act. They can be considered as active securities, which go hand-in-hand with worker’s initiatives to shoulder the risks of flexible employment relationships instead of restricting them.

Conclusion The intrusion of ‘risk management’ into the public domain will probably keep the discourse on employment and social policy lively for more than 9 years. The main reason for this is that the present labour market institutions, especially unemployment insurance, no longer provide adequate security, particularly with respect to the new risks mentioned in the second part of this paper. The present 19

The ILO-concept of ‘decent work’ corresponds to this notion: ‘A decent society—not a utopia—is one committed to the extension of real freedom, which is a combination of what Isaiah Berlin called negative liberty and positive liberty. This means that individuals and social groups should have basic economic security, for otherwise they cannot be expected to be able to make rational decisions and good opportunity for the pursuit of occupation. The former requires at least a minimum degree of income security coupled with basic voice security, i.e., an assurance of a stable and decent level of remuneration coupled with the assurance that there will be both collective and individual representation on case of dispute or bargaining’. (Standing, 2003, p. 12).

28

G. Schmid

system especially does not provide adequate incentive structures to prevent or mitigate potential unemployment through voluntary mobility, as the (TLM) theory recommends. The notion of ‘making transitions pay’20 through new social rights has not yet fully entered the world of work. The same can be said for ‘making flexible work organization pay’, its counterpart on the demand side. However, new securities are a precondition for encouraging riskier decisions such as moving from dependent employment to self-employment or even embarking on the venture of changing occupation once or may be twice in the life course. Such ‘internal’ or ‘self-generated’ risks are increasing anyhow, as shown in the risk analysis. What is more, they are increasingly desired in order to improve firms’ adaptability to changing conditions or even to realize changes in individual life circumstances or preferences. The old insurance systems still contain many elements that reflect the traditional gender division. Internal labour markets, which are essentially insurance devices, are being eroded, and the open-ended employment contract is in crisis. However, functionally equivalent systems offering insurance against earnings risks over the life course are not yet in place or only just beginning to emerge. For this reason, some form of employability insurance should supplement the time-honoured system of unemployment insurance. This should provide cover for income risks not only in the worst case, that is unemployment, but also during risky transitions into new employment relationships. The various elements of such an employability insurance system, e.g. wage insurance, time accounts, training or learning accounts, the transformation of benefit entitlements into vouchers, entitlements or ‘social drawing rights’ to training or education leaves or sabbaticals have been described elsewhere.21 The extension of the employability concept, especially embedded employability, also requires the social partners to take a more active and innovative role in the formation and control of such additional insurance elements.22 As our outline of prospect and framing theory has shown, the establishment of a reliable opportunity set through TLMs would overcome (at least to some extent) non-rational risk aversion leading to suboptimal employment careers and inflexible as well as segmented labour markets. Extending the idea of insurance to income volatility over the life course would increase overall mobility and thereby not only open up the doors to the long-term unemployed but also

20

This slogan was coined by Bernard Gazier.

21

Cf. Schmid (2002), Schmid and Gazier (2002, pp. 398–435), Supiot (2001) and Wilthagen and Rogowski (2002).

22

David Marsden (2003) has emphasized this point also from a historical point of view showing, for instance, that the British trade unions played a pioneer role in mutual insurance.


29

reduce the inflow into unemployment by preventing or mitigating risks to a much greater extent for both employers and employed insiders. More flexibility with reasonable levels of security is indeed the aim of TLMs. However, it would be a mistake to see flexibility and security only as a trade-off and to see the future of work only as a problem of achieving the appropriate balance between flexibility and security. Rather, the aim of social risk management through TLMs is the more ambitious one of developing a complementary relationship between flexibility and security.

Acknowledgements I wish to thank especially two anonymous referees; unfortunately, I was not able to respond to all of their excellent suggestions, partly due to limitations of space, partly because some points need further research and reflections. I also address my gratitude to Je´roˆme Gautie´, Anton Hemerijck, Achim Kemmerling, Janine Leschke and Michael Neugart for comments on earlier versions, to Dorit Griga, Brigitte Zarek and Jutta Ho¨hne for invaluable assistance and to Andrew Wilson for checking and improving on my English.

References Arjona, R., Ladaique, M. and Pearson, M. (2001) ‘Growth, Inequality and Social Protection, Paris’, OECD Labour Market and Social Policy Occasional Papers Number 51, DEELSA/ELSA/WD(2001)4. Arrow, K. J. (1974) Essays in the Theory of Risk-Bearing, Amsterdam and Oxford, NorthHolland Publishing Company; New York, American Elsevier Publishing Company. Barbier, J. C. (2004) ‘Activation Policies: A Comparative Perspective’. In Serrano Pascual, A. (ed) Are Activation Policies Converging in Europe? The European Employment Strategy for Young People, Brussels, ETUI. Beck, U. [1986] (1992) Risk Society: Towards a New Modernity, London, Sage. Begg, I. and Berghman, J. (2002) ‘Introduction: EU Social (Exclusion) Policy Revisited?’, Journal of European Social Policy, 12, 179–94. Bernstein, P. L. (1996) Against the Gods. The Remarkable Story of Risk, New York, John Wileys & Sons. Boeri, T., Boersch-Supan, A. and Tabellini, G. (2001) ‘Would you Like to Shrink the Welfare State? A Survey of European Citizens’, Economic Policy, 32, 9–50. Breyer, F., Franz, W., Homburg, S., Schnabel, R. and Wille, E. (2004) Reform der sozialen Sicherung, Heidelberg, Springer. de Bruijn, J., de Vries, G. and ten Cate, H. (2003) ‘The Burden of Late Modernity: Work-Related Disability Among Women in The Netherlands’, unpublished manuscript.

30

G. Schmid

Considine, M. (2001) Enterprising States. The Public Management of Welfare-to-Work, Cambridge, Cambridge University Press. Daly, M. (2003) ‘Governance and Social Policy’, Journal of Social Policy, 32, 113–28. Dworkin, R. (2000) Sovereign Virtue. The Theory and Practice of Equality, Cambridge, MA and London, Harvard University Press. Esping-Andersen, G., Gallie, D., Hemerijck, A. and Myles, J. (2002) Why We Need a New Welfare State, Oxford, Oxford University Press. European Commission (2002) [2003, 2004] Employment in Europe 2002 [2003, 2004]— Recent Trends and Prospects, Luxembourg, Office for Official Publications of the European Communities. Feldstein, M. (2005) ‘Rethinking Social Insurance’, American Economic Review, 95, 1–24. Fo¨lster, S. (1999) ‘Social Insurance Based on Personal Savings’, The Economic Record, 75, 5–18. Frank, R. H. and Cook, P. J. (1995) The Winner-Take-All Society, New York, The Free Press. Gangl, M. (2003) Unemployment Dynamics in the United States and West Germany. Economic Restructuring, Institutions and Labour Market Processes, Heidelberg and New York, Physica/Springer. Gautie´, J. (2004) ‘Les marche´s internes du travail, l’emploi et les salaires,’ Revue Fran¸caise d’Economie, 18, 33–62. Gazier, B. (1999) ‘Definitions and Trends’. In Gazier, B. (ed) Employability: Concepts and Policies, Employment Observatory Research Network, Report 1998, Berlin, I.A.S. Gazier, B. (2003) ‘Tous Sublimes’. Vers un nouveau plein-emploi, Paris, Flammarion. Giddens, A. (1998) The Third Way. The Renewal of Social Democracy, Cambridge, Polity Press. Giddens, A. (2003) ‘Introduction. The Progressive Agenda’. In Browne, M., Thompson, P. and Sainsbury, F. (eds) Progressive Futures. New Ideas for the Centre-Left, London, Policy Networks. Gigerenzer, G. (2002) Reckoning with Risk: Learning to Live with Uncertainty, London, Penguin Books. Heimer, C. A. (2003) ‘Insurers as Moral Actors’. In Ericson, R. V. and Doyle, A. (eds) Risk and Morality, Toronto, Buffalo and London, University of Toronto Press. Hirschmann, A. O. (1982) Shifting Involvements. Private Interest and Public Action, Princeton, NJ, Princeton University Press. Hoffmann-Riem, W. (2001) Modernisierung von Recht und Justiz. Eine Herausforderung des Gewa¨hrleistungsstaates, Frankfurt a.M., Suhrkamp. Holzmann, R. and Jorgensen, S. (2000) ‘Social Risk Management: A New Conceptual Framework for Social Protection and Beyond’. Social Protection Discussion Paper Series No. 0006, Washington, The World Bank.


31

Kahnemann, D. (2003) ‘Maps of Bounded Rationality: A Perspective of Intuitive Judgment and Choice’. In Frangsmyr, T. (ed) Les Prix Nobel 2002, Stockholm, Imprimerie Royal, P.A. Norstedt & So¨ner (http://www.nobel.se/economics/laureates/ 2002/kahnemann-lecture.pdf). Kahnemann, D. and Tversky, A. (eds) (2000) Choices, Values, and Frames, Cambridge, Cambridge University Press. Keller, H. E. (2004) ‘Auf sein Aventura und Risigo handeln’. Zur Sprach- und Kulturgeschichte des Risiko-Begriffs’, Risknews, 1, 61–5. de Koning, J. and Mosley, H. (eds) (2001) Labour Market Policy and Unemployment. Impact and Process Evaluations in Selected European Countries, Cheltenham, UK and Northampton, MA, Edward Elgar. Le Grand, J. (2003) Motivation, Agency, and Public Policy. Of Knights and Knaves, Pawns and Queens, Oxford, Oxford University Press. Lewis, J. and Guillari, S. (2005) ‘The Adult Worker Model Family, Gender Equality and Care: The Search for New Policy Principles and the Possibilities and Problems of a Capabilities Approach’, Economy and Society, 34, 76–104. Luhmann, N. (1990) Die Wissenschaft der Gesellschaft, Frankfurt a.M., Suhrkamp. Macdonald, F. and Holm, S. (2001) ‘Employment for 25- to 34-Year-Olds in the Flexible Labour Market: A Generation Excluded?’. In Hancock, L., Howe, B., Frere, M. and O’Donnell, A. (eds) Future Directions in Australian Social Policy. New Ways of Preventing Risk, Melbourne, CEDA. Mangan, J. (2000) Workers Without Traditional Employment: An International Study of Non-Standard Work, Cheltenham, UK, and Northampton, MA, Edward Elgar. Marsden, D. (2003) ‘Adapting European Labour Institutions in Global Economic and Technical Change’. In Touffut, J.-Ph. (ed) Institutions, Innovation and Growth, Selected Economic Papers, Cheltenham, UK and Northampton, MA, USA, Edward Elgar. Meager, N. and Bates, P. (2002) ‘From Salary Workers to Entrepreneurial Workers?’ In Schmid, G. and Gazier, B. (eds) The Dynamics of Full Employment. Social Integration Through Transitional Labour Markets, Cheltenham, UK and Northampton, MA, Edward Elgar. Mosley, H., O’Reilly, J. and Scho¨mann, K. (eds) (2002) Labour Markets, Gender and Institutional Change. Essays in Honour of Gu¨nther Schmid, Cheltenham, UK and Northampton, MA, Edward Elgar. Muffels, R., Wilthagen, T. and van den Heuvel, N. (2002) Labour Market Transitions and Employment Regimes: Evidence on the Flexicurity-Security Nexus in Transitional Labour Markets, WZB Discussion Paper FS I 02–204, Berlin, Wissenschaftszentrum Berlin fu¨r Sozialforschung. Neugart, M. and Scho¨mann, K. (eds) (2002) Forecasting Labour Markets in OECD Countries, Cheltenham, UK and Northampton, MA, Edward Elgar.

32

G. Schmid

Nussbaum, M. (2000) Women and Human Development. The Capabilities Approach, Cambridge, Cambridge University Press. Nussbaum, M. (2003) ‘Capabilities as Fundamental Entitlements: Sen and Social Justice’, Feminist Economics 9, 33–59. O’Reilly, J., Cebria´n, I. and Lallement, M. (eds) (2000) Working-Time Changes. Social Integration Through Transitional Labour Markets, Cheltenham, UK and Northampton, MA, Edward Elgar. Rawls, J. (1990) A Theory of Justice, Oxford, Oxford University Press. Rawls, J. (2001) Justice as Fairness – A Restatement, Cambridge and London, The Belknap Press of Harvard University Press. Salais, R. (2003) ‘Work and Welfare: Toward a Capability Approach’. In Zeitlin, J. and Trubek, D. M. (eds) Governing Work and Welfare in a New Economy. European and American Experiments, Oxford, Oxford Unviersity Press. Sarfati, H. and Bonoli, G. (eds) (2002) Labour Market and Social Protection Reforms in International Perspective. Parallel or Converging Tracks? Aldershot, England and Burlington, USA, Ashagete. Schmid, G. (1993) ‘Equality and Efficiency in the Labor Market. Towards a SocioEconomic Theory of Cooperation in the Globalizing Economy’, The Journal of SocioEconomics, 22, 31–67. Schmid, G. (1995) ‘Is Full Employment Still Possible? Transitional Labour Markets as a New Strategy of Labour Market Policy’, Economic and Industrial Democracy, 16, 429–56. ¨ bergangsarbeitsma¨rkte und Schmid, G. (2002) Wege in eine neue Vollbescha¨ftigung. U aktivierende Arbeitsmarktpolitik, Frankfurt a.M., Campus. Schmid, G. (2003) ‘Activating Labour Market Policy: ‘Flexicurity’ Through Transitional Labour Markets’. In Touffut, J.-Ph. (ed) Institutions, Innovation and Growth, Cheltenham, UK and Northampton, MA, USA, Edward Elgar. Schmid, G. and Gazier, B. (eds) (2002) The Dynamics of Full Employment. Social Integration Through Transitional Labour Markets, Cheltenham, UK and Northampton, MA, Edward Elgar. Scho¨mann, K. and O’Connell, Ph. J. (eds) (2002) Education, Training and Employment Dynamics. Transitional Labour Markets in the European Union, Cheltenham, UK and Northampton, MA, Edward Elgar. Scho¨mann, K., Rogowski, R. and Kruppe, T. (1998) Labour Market Efficiency in the European Union. Employment Protection and Fixed-Term Contracts, London and New York, Routledge. Schuppert, G. F. (2003) Staatswissenschaften, Baden-Baden, Nomos. Sen, A. (2001) Development as Freedom, New York, Alfred A. Knopf. Shiller, R. (2003) The New Financial Order: Risk in the 21st Century, Princeton, Princeton University Press.


33

Sinn, H.-W. (1995) ‘Social Insurance, Incentives and Risk Taking’, Working Paper 5335 (NBER Working Paper Series), Cambridge, MA, National Bureau of Economic Research. Solow, R. M. (1990) The Labor Market as a Social Institution, Oxford, Basil Blackwell. Standing, G. (1999) Global Labour Flexibility. Seeking Distributive Justice, London and New York, Macmillan and St Martin’s Press. Standing, G. (2003) The Decent Work Enterprise: Worker Security and Dynamic Efficiency, International Labour Office, Geneva. Available at http://www.ilo.org/ses Stiglitz, J. and Yun, J. (2002) ‘Integration of Unemployment Insurance with Retirement Insurance’, NBER Working Paper Series, Working Paper 9199. Available at http:// www.nber.org/papers/w9199 Supiot, A. (2001) Beyond Employment. Changes in Work and the Future of Labour Law in Europe, Oxford, Oxford University Press. Taylor-Gooby, P. (ed) (2004) New Risks, New Welfare. The Tranformation of the New Welfare State, Oxford, Oxford University Press. Vandenbroucke, F. (2001) Social Justice and Individual Ethics in an Open Society. Equality, Responsibility, and Incentives, Berlin and New York, Springer. Williamson, O. E. (1996) The Mechanisms of Governance, New York and Oxford, Oxford University Press. Wilthagen, T. and Rogowski, R. (2002) ‘Legal Regulation of Transitional Labour Markets: Barriers and Opportunities’. In Schmid, G. and Gazier, B. (eds) The Dynamics of Full Employment. Social Integration Through Transitional Labour Markets, Cheltenham, UK and Northampton, MA, Edward Elgar.